Leak detection method with time delay for detecting a leak that affects a tyre curing membrane or the inflation circuit thereof

Abstract

The invention relates to a method for detecting a leak in a curing membrane within a facility (1) for curing tyres (2) in a mould (3), the method comprising, after opening the mould (3), a test step (S200) which comprises a sub-step (S201) of partial re-inflation during which the curing membrane (10) is reinflated and, when the internal pressure prevailing in the curing membrane reaches a predetermined threshold, referred to as the "test pressure" (P test), the curing membrane (10) is isolated and a predetermined time delay (d_202) is allowed to elapse, at the end of which the value of the internal pressure, referred to as the "reference pressure" (P_ref), is measured, followed by a sub-step (S203) of observation during which changes to the internal pressure relative to the reference pressure (P_ref) are observed in order to identify a leak situation.

Description

LEAK DETECTION METHOD WITH TIME DELAY FOR DETECTING A LEAK AFFECTING A BANDAGE COOKING MEMBRANE OR THE INFLATION CIRCUIT OF SAID MEMBRANE

[0001] The present invention relates to the field of manufacturing tires intended to equip vehicle wheels, in particular pneumatic tires.

[0002] It relates more particularly to the field of the operation of curing bandages, which makes it possible to give a bandage its final appearance, in particular by shaping the sculptures of the tread of said bandage, and to vulcanize said bandage.

[0003] It is known, to carry out this curing operation, to place a blank of bandage, or "raw bandage", unvulcanized, or at least not entirely vulcanized, in a curing mold whose internal wall corresponds to the negative imprint of the bandage that it is desired to obtain, then to close said curing mold, then to force the expansion of an elastic curing membrane, located inside the mold, between the beads of the bandage, by means of a pressurized inflation fluid, such as water vapor, in such a way that the external surface of said curing membrane comes to bear against the internal surface of the bandage and presses the latter against the internal wall of the mold. The heat provided by the mold and by the inflation fluid allows the vulcanization of the bandage.After the baking operation, the membrane is deflated, the mold is opened, and the baked bandage is removed to be replaced by a new raw bandage, in preparation for a new baking operation.

[0004] The thermal and mechanical conditions of repeated use of the cooking membrane are particularly severe, and are therefore the cause of damage to the cooking membrane, in particular through the appearance of perforations which cause a loss of sealing of the cooking membrane, and which may, in certain cases, be the cause of a tear in said cooking membrane during a cooking operation.

[0005] Of course, such a failure of the curing membrane generally leads to the scrapping of the bandage, either because said bandage will have been insufficiently cured due to the interruption of the curing operation, or because said bandage will have been poorly shaped due to the impossibility of putting the membrane under pressure, or because said bandage will have been damaged and therefore weakened by an intrusion of the inflation fluid which will have leaked out of the curing membrane and which will have insinuated itself inside the very structure of the bandage, thus causing decohesion between the layers of rubber constituting said bandage.

[0006] However, since damage to the cooking membrane is a progressive process, although highly variable from one membrane to another, it was considered to detect possible leaks affecting the cooking membrane in order to allow preventive, but nevertheless not premature, replacement of the cooking membrane, before a tear in said cooking membrane occurs.

[0007] Thus, document EP-1 213 573 discloses in particular a method for detecting a leak in a cooking membrane according to which, after the cooking operation, the cooking mold is partially opened while the cooking membrane is still under pressure, so that, if the cooking membrane is perforated, part of the pressurized water vapor contained in the cooking membrane can escape through the perforation and force a passage between the external surface of the cooking membrane and the internal surface of the bandage, until it reaches a humidity sensor which is placed close to the central axis of the mold.

[0008] However, such a known method only allows very imperfect and rather unreliable detection of leaks, since, for detection to be effective, it is necessary, on the one hand, that the orifice at the origin of the leak is large enough to allow a sufficient quantity of steam to escape so that the resulting variation in humidity in the vicinity of the humidity sensor is perceptible by said humidity sensor, on the other hand, that the cooked bandage, which is still in the mold, does not block said leak orifice too firmly so that the steam can force the passage between the membrane and the bandage, and finally that the passage that the steam creates between the cooking membrane and the bandage actually conducts the flow of steam to the humidity sensor, without dispersing said flow within of the mold nor divert said steam flow to another exhaust port which would be distant from said humidity sensor.

[0009] Furthermore, such a process is potentially dangerous, since it requires opening the mold while the curing membrane is still under high pressure, because high pressure is indeed required to generate the leakage flow and force a passage of water vapor between the curing membrane and the bandage. This therefore increases, during this critical phase, the risk of sudden tearing, or even bursting, of the curing membrane, damage to the bandage, and possibly projection of hot steam and therefore bodily harm to an operator who is near the mold.

[0010] Furthermore, this same document indicates that a leak detection system which would no longer rely on a humidity measurement but rather on a pressure measurement would not be reliable, and in particular would be incapable of detecting small perforations, because the leak rate, and therefore the corresponding variation in pressure, would then be so low that they would be undetectable in practice.

[0011] The inventors were also able to confirm for themselves this problem of the undetectability of small leaks by known methods, by noting that, during leak tests carried out by them in accordance with these known methods, condensation phenomena of the water vapor used to inflate the membrane occur, and that these condensation phenomena cause, inside the membrane, pressure drops whose amplitude is of the same order of magnitude as, or even greater than, the amplitude of the pressure variations attributable to the leak itself. It is then impossible to distinguish between a leak and a normal condensation phenomenon whose effects mask those of said leak.

[0012] Furthermore, it may happen that a cooking membrane suffers from a leak which is sufficient to cause damage to the bandage, in particular by allowing enough inflation fluid to escape to cause decohesion of the constituent layers of the bandage, but which is nevertheless too small to be detected directly at the level of the cooking installation. In such a situation, the failure of the cooking installation may only become apparent much later, i.e. directly on the cooking installation itself, when the leak finally causes a clearly visible tear in the cooking membrane, or indirectly, downstream of the cooking installation, when it is found, at a quality control installation for finished bandages, that the bandages coming from the cooking installation have defects. However, a detrimental consequence of the late detection of the membrane defect is that the cooking installation is likely to produce, during the time between the appearance of the leak and the detection of the defect, a whole batch of defective bandages, before the leak in the cooking membrane is finally diagnosed and the defective membrane replaced. This of course results in a waste of raw materials and energy, and therefore a loss of industrial efficiency.

[0013] The objects assigned to the invention therefore aim to remedy the aforementioned drawbacks and to propose a new method for detecting leaks within a bandage curing installation, and more particularly for detecting leaks affecting a curing membrane, in a particularly reliable and early manner, even when these leaks result from small perforations of the curing membrane, and this quickly, in order to have the smallest possible impact on the useful time of said curing installation.

[0014] The objects assigned to the invention are achieved by means of a leak detection method intended to detect a leak within a cooking installation for cooking bandages, said installation comprising: - a mold which is arranged to alternately adopt a closing configuration, in which the internal wall of said mold, which corresponds to the negative imprint of the bandage, delimits a molding cavity and allows a baking operation of said bandage, and an opening configuration which allows, after the baking operation, a baked bandage to be extracted from said mold and replaced by a raw bandage, - a cooking membrane whose internal surface delimits an inflation chamber allowing the cooking membrane to be inflated, and whose external surface is arranged so as, during the cooking operation, and under the action of a pressure called "cooking pressure" which is applied in the inflation chamber, to come to bear against the internal surface of the bandage in order to press the bandage against the internal wall of the mold, - an inflation circuit which comprises at least one inlet valve allowing a fluid to be introduced into the cooking membrane, into the inflation chamber pressure inflation to inflate the cooking membrane, - a pressure measuring device arranged to measure the pressure known as “internal pressure”, which prevails inside the cooking membrane, in the inflation chamber, said method comprising: - a step (S 100) of opening the mold, during which, after deflating the cooking membrane, the mold is placed in the opening configuration, - a test step (S200) which comprises: a partial re-inflation sub-step (S201), during which, while the mold is in the open configuration, the opening of the inlet valve is controlled in order to introduce inflation fluid into the inflation chamber to increase the internal pressure prevailing in said inflation chamber and thus cause re-inflation of the cooking membrane, while measuring said internal pressure prevailing in said inflation chamber by means of the pressure measuring device, then a stabilization sub-step (S202) which is triggered when the internal pressure prevailing in the inflation chamber reaches a predetermined threshold, called "test pressure", which is chosen to be lower than the cooking pressure, and according to which on the one hand the inflation chamber is isolated by controlling the closing of the inlet valve, and on the other hand a time delay of a predetermined duration is applied,called "delay period", and, at the end of said delay period, the value called "reference pressure" of the internal pressure prevailing in the inflation chamber is measured, then an observation sub-step (S203), during which one observes how the internal pressure prevailing in the inflation chamber evolves over a predetermined duration called "observation period" (d_203), in order to identify a change in the internal pressure, compared to the reference pressure, which is representative of a leak situation.,

[0015] Advantageously, the implementation of a time delay according to the invention, triggered at the moment when the chosen test pressure is reached and the closing of the inlet valve is ordered, makes it possible to ensure a certain stabilization of the system, and thus to eliminate transient phenomena which cause pressure variations likely to disrupt the detection of a leak. These transient phenomena may in particular be linked to: - to the inertia of the intake valve, which inertia is likely to cause a overshoot of the effective internal pressure value compared to the chosen test pressure, especially since the increase in internal pressure during the re-inflation phase generally has particularly rapid dynamics, and / or - possible pressure oscillations linked to the mechanical reactions of the inlet valve, the compressibility of the inflation fluid and / or the elastic or visco-elastic behavior of the cooking membrane.

[0016] Thus, the invention makes it possible to detect a pressure variation indicative of a leak in a very precise manner, after the attenuation or even the extinction of any transient disturbing phenomena. The possible residual amplitude of said possible transient disturbing phenomena is consequently sufficiently low, or even zero, at the time when the measurement of the reference pressure is carried out at the end of the time delay / stabilization phase, and then subsequently during the actual observation phase, so that said possible residual amplitude of the transient disturbing phenomena on the one hand does not distort the value of the reference pressure, and on the other hand does not mask the detection of a pressure variation, even a small one, which is actually attributable to a leak. The leak detection method according to the invention thus has a reliability and sensitivity much higher than those of known methods.

[0017] Thanks to this increased sensitivity, the method according to the invention advantageously makes it possible to detect a minor leak very early on, well before it presents a real risk of damaging the bandages or tearing the cooking membrane.

[0018] Furthermore, advantageously, the control of the respective durations of delay and then observation, which remain relatively reduced with respect to the cooking cycle time and in particular with respect to the time necessary to extract the cooked bandage from the mold and place a new raw bandage in said mold to initiate a new cooking cycle, makes it possible to carry out a leak test quickly. Thus, the implementation of the leak detection method according to the invention immobilizes the cooking installation little, if any, and has little, if any, harmful consequences on the useful time of said cooking installation, and therefore on the average cooking cycle time, considered over several successive operations.

[0019] Moreover, it will be noted that the gain in reliability provided by the method according to the invention makes it possible to significantly reduce the occurrence of leaks damaging to the bandages, and thus to limit in the long term the rate of bandages rendered non-compliant by such leaks, which amply compensates, in terms of industrial efficiency, for a possible slight extension of the average cooking cycle time.

[0020] Finally, advantageously, as will be detailed below, the method according to the invention makes it possible to verify, in a single test, the tightness of the cooking membrane and the correct operation of the inlet valve.

[0021] Other objects, characteristics and advantages of the invention will appear in more detail on reading the description which follows, as well as with the aid of the appended drawings, provided for purely illustrative and non-limiting purposes, among which:

[0022] Figure 1 illustrates, in a sectional view in a plane containing the central axis of the mold, here a vertical plane, a cooking mold in the closed configuration, during a cooking operation, the cooking membrane, here shown in dotted lines, being placed under the cooking pressure and pressing the bandage against the internal wall of the mold.

[0023] Figure 2 is a sectional view, in the same sectional plane as that of Figure 1, of the cooking mold in the open configuration, the cooking membrane being deflated, and more particularly placed under depression so as to reduce its overall volume, which results in a retraction of said membrane by pleating, to allow the removal of the cooked bandage, just before the test step aimed at detecting a possible leak.

[0024] Figure 3 is a sectional view, in the same sectional plane as that of Figures 1 and 2, of the cooking mold in the open configuration, during a first phase of the re-inflation sub-step of the test step, the cooking membrane being in an intermediate re-inflation configuration which corresponds to its rest configuration, that is to say to the configuration adopted by said membrane when the internal pressure prevailing in the inflation chamber is equal to the ambient atmospheric pressure which prevails around, and is applied to, the external surface of the membrane.

[0025] Figure 4 is a sectional view, in the same sectional plane as that of Figures 1 to 3, of the cooking mold in the opening configuration, during a second phase of the re-inflation sub-step of the test step, the cooking membrane being partially re-inflated, under an internal pressure equal to the desired test pressure, higher than the ambient atmospheric pressure and lower than the cooking pressure, and said cooking membrane being isolated by closing the supply valve to allow the stabilization and then observation sub-steps intended to detect a possible leak to take place.

[0026] Figure 5 is an overall perspective view of a cooking membrane according to the invention, in the resting configuration.

[0027] Figure 6 is a graph illustrating, schematically, the evolution of the internal pressure prevailing in the inflation chamber, and therefore more generally in the cooking membrane, as a function of time, during a test step according to the invention, and showing in particular the different sub-steps of partial re-inflation, stabilization by time delay and then observation. On this graph, the scenario represented by a solid line during the observation sub-step reflects a small drop in pressure which corresponds to an absence of leakage, the scenario represented by a dashed line, which reflects a significant drop in pressure during the observation period, corresponds to a situation of leakage of the cooking membrane, and the scenario represented by a dotted line, which reflects an increase in pressure during the observation period, corresponds to a situation of leakage of the intake valve.

[0028] Figure 7 is a graph illustrating the evolution of the pressure prevailing in the cooking membrane during a sequence comprising first of all a test step according to the invention, then a cooking operation of a bandage.

[0029] The present invention relates to a leak detection method intended to detect a leak within a cooking installation 1 for cooking bandages 2.

[0030] Said bandages 2 are intended to equip vehicle wheels, and may in particular constitute pneumatic bandages, within which the wall of the bandage defines, with the rim on which the bandage is mounted, a chamber filled with a pressurized gas which supports said bandage, or else so-called “airless” bandages which are supported by a network of spokes rather than by a chamber filled with pressurized gas. The bandages 2 may also, as a variant, constitute tracks.

[0031] Preferably, said bandages 2 comprise one or more rubber-based components.

[0032] All or part of these rubber-based components, and therefore more generally the bandage 2, are initially in the raw state, that is to say unvulcanized, and the curing installation 1 advantageously makes it possible to implement a curing operation intended to vulcanize said rubber-based component(s), to obtain a cured bandage.

[0033] As a guide, the cooking operation generally takes place at a temperature between 130°C and 200°C.

[0034] As can be seen in Figures 1, 2 and 3, the installation comprises a mold 3 which is arranged to alternately adopt a closing configuration (Figure 1), in which the internal wall 3_in of said mold 3, which corresponds to the negative imprint of the bandage 2, delimits a molding cavity and allows a baking operation of said bandage 2, and an opening configuration (Figures 2, 3 and 4) which allows, after the baking operation, a baked bandage 2 to be extracted from said mold 3 and to replace it with a raw bandage 2.

[0035] In a manner known per se, the mold 3 has a central axis Z3, here vertical, with respect to which the mold 3 has a substantially revolutionary shape, and which corresponds to the central axis of the bandage 2 to be baked.

[0036] The mold 3 also comprises plates 4, 5, here a lower plate 4 and an upper plate 5, which are movable relative to each other along the central axis Z3, so as to be able to alternately move towards each other to ensure the closing of the mold 3, and move away from each other to ensure the opening of the mold 3.

[0037] The opening and closing of mold 3 takes place according to a joint plane noted P3.

[0038] The mold 3 comprises a set of blocks 6, called “sectors” 6, which each cover a predetermined angular sector around the central axis Z3, and which together form an annular structure whose internal wall, which forms part of the internal wall 3_in of the mold, corresponds to the negative imprint of the top of the bandage 2. In a manner known per se, the sectors 6 are equipped with slats which project towards the inside of the molding cavity and whose design corresponds to that of the sculptures of the tread of the bandage 2.

[0039] The mold 3 preferably further comprises an annular hoop 7 which is axially movable along the central axis Z3 and which cooperates along a frustoconical ramp 7A with the sectors 6, so as to be able to alternately force the centripetal radial approach of the sectors 6 when the mold 3 is closed, then force the centrifugal radial separation of said sectors when the mold 3 is opened.

[0040] The mold 3 also comprises shells 8, 9 which are carried by the plates 4, 5 and of which an internal wall, which forms another part of the internal wall 3_in of the mold, makes it possible to shape the sides of the bandage, and to affix certain markings there, in particular regulatory markings.

[0041] The installation 1, and more particularly the mold 3, also comprises a cooking membrane 10 whose internal surface 10_in delimits an inflation chamber 11 which makes it possible to inflate the cooking membrane 10, and whose external surface 10 out is arranged to, during the cooking operation, and under the action of a pressure called "cooking pressure" P eur which is applied in the inflation chamber 11, come to bear against the internal surface 2_in of the bandage 2 in order to press the bandage 2, and more particularly the external surface 2_out of the bandage 2, against the internal wall 3_in of the mold 3, as illustrated in FIG. 1.

[0042] The cooking membrane 10, in a manner known per se, preferably has, in particular at rest, in the absence of internal overpressure relative to the ambient pressure, a shape of revolution around the central axis Z3, either straight cylindrical, the generatrices of the membrane being parallel to the central axis Z3, or curved in the shape of a barrel, as illustrated in FIG. 5, the generatrices of the membrane then being curved relative to the central axis Z3.

[0043] The cooking membrane 10 is made of an elastically deformable material, such as vulcanized rubber, to allow the expansion of said cooking membrane, and therefore the expansion of the volume of the inflation chamber 11, under the effect of the pressure exerted in said inflation chamber 11.

[0044] The thickness of the cooking membrane 10, which corresponds to the distance which separates the external surface 10 out from the internal surface 10_in, is generally, at rest, between 4 mm and 15 mm, and for example between 4 mm and 6 mm when the installation is intended for the curing of tires intended for passenger vehicles, generally up to 10 mm for the curing of tires intended for heavy goods vehicles, and beyond for the curing of tires intended for civil engineering machinery.

[0045] The cooking membrane 10 has, at each of its axial ends, an annular bulge forming a heel 12, 13 which is held in a sealed manner by a flange 14, 15.

[0046] The installation 1 further comprises an inflation circuit 20 which comprises at least one inlet valve 21 making it possible to introduce into the cooking membrane 10, in the inflation chamber 11, an inflation fluid under pressure in order to inflate the cooking membrane 10.

[0047] The inflation circuit 20 also comprises at least one exhaust valve 22 allowing the inflation fluid to be evacuated from the inflation chamber 11 in order to deflate the cooking membrane 10.

[0048] According to a possible alternative embodiment, and as illustrated in Figures 1 and 2, the inlet valve 21 and the exhaust valve 22 can be combined into a single valve, which advantageously forms an isolation valve, the communication between the inflation circuit 20 and the inflation chamber 11 being effected by a single pipe 23 which carries said isolation valve and which serves in turn for the admission then the exhaust of the inflation fluid, and therefore for the inflation then the deflation of the cooking membrane 10.

[0049] According to another variant embodiment, the intake valve 21 and the exhaust valve 22 are separate, the intake and exhaust being carried out by two separate pipes which each communicate with the inflation chamber 11, namely a first pipe carrying the intake valve 21 and a second pipe carrying the exhaust valve 22.

[0050] In a manner known per se, the inflation circuit 20 comprises a compressor and / or an accumulator (not shown) which makes it possible to supply the inflation chamber 11 with the inflation fluid under pressure, via the inlet valve 21.

[0051] By "under pressure", we conventionally indicate that the inflation fluid is under pressure relative to the ambient pressure, and more particularly to the ambient atmospheric pressure P atm, which prevails in the area where the installation 1 is located, and more particularly which prevails around the mold 3.

[0052] According to a preferred embodiment, the inflation circuit 20 may also comprise a suction system, such as a vacuum pump, making it possible to temporarily place the inflation chamber 11 under vacuum, that is to say under an internal pressure which is lower than the ambient pressure which prevails around the mold 3, and therefore in particular which is lower than the ambient pressure which prevails at the external surface 10 out of the cooking membrane 10, when the mold 3 is in the open configuration, and this in order to retract the cooking membrane 10 on itself, as illustrated in FIG. 2, and thus prevent the cooking membrane 10 from interfering with the bandage 2, and in particular with the heels of said bandage 2, when said bandage 2 is extracted from the mold 3 after cooking or when a new raw bandage 2 is introduced into said mold 3.

[0053] The installation 1 also comprises a pressure measuring device 25 arranged to measure the pressure, called “internal pressure”, which prevails inside the cooking membrane 10, in the inflation chamber 11.

[0054] This pressure measuring device 25 may comprise a pressure sensor 26, of any suitable type, which pressure sensor 26 may for example be located on a portion of the pipe 23 which communicates permanently with the inflation chamber 11, here a portion of the pipe 23 between the intake valve 21 and the inflation chamber 11.

[0055] The installation 1 further advantageously comprises a control unit 27, preferably electronic, such as a programmable controller, which makes it possible to regulate the internal pressure which prevails within the cooking membrane 10, in the inflation chamber 11.

[0056] Said control unit 27 is advantageously connected to the pressure measuring device 25, and makes it possible to control, preferably automatically, the inflation circuit 20, and more particularly the opening and closing of the intake valve 21 and the exhaust valve 22, as a function of a pressure setpoint or a law of chosen regulation. The intake valve 21 and the exhaust valve 22 may advantageously be solenoid valves, controlled by electrical control signals emitted by the control unit 27.

[0057] Thus, the control unit 27 makes it possible to put the inflation chamber 11 in communication sometimes with the intake circuit of the inflation circuit 20, and therefore with the compressor or the accumulator, in order to increase the internal pressure by injecting inflation fluid under pressure into the inflation chamber 11, sometimes with the exhaust circuit of the inflation circuit 20 which allows a reduction of the internal pressure by allowing inflation fluid to escape from the inflation chamber, or even which generates a forced emptying of the inflation fluid by creating a vacuum in said inflation chamber 11.

[0058] The control unit 27 also makes it possible, as will be seen below, to isolate the inflation chamber 11 from the inflation circuit 20, so that said inflation chamber no longer communicates with either the intake circuit or the exhaust circuit, in order to allow the internal pressure, which prevails in said inflation chamber 11 thus separated from the inflation circuit 20, to evolve freely, passively, without being subject to active regulation. In other words, the control unit 27 can, when necessary, inhibit the pressure regulation, in particular, as will be seen below, to be able to detect an abnormal variation in the internal pressure which would be indicative of a leak.

[0059] According to the invention, the method firstly comprises a step (S100) of opening the mold 3, during which, after having deflated the cooking membrane 10, the mold 3 is placed in the opening configuration, as illustrated in FIG. 2.

[0060] To do this, the hoop 7, here carried by the upper plate 4, is moved in order to radially separate the sectors 6 so as to release said sectors 6 beyond the radial limits of the cured bandage 2, then the upper plate 4 is axially lifted, which carries with it the hoop 7 and the separated sectors 6, so as to expose the lower plate 5, the bandage 3 which rests on said lower plate 5, and the curing membrane 10 which is in the deflated state, here more particularly in a retracted state due to a vacuum.

[0061] It is thus possible to access the cooked bandage 2 and remove it from the mold 3, by sliding said cooked bandage 2 axially, along the cooking membrane 10, by means of a suitable manipulator, such as a robotic arm.

[0062] It will be noted that, during this step (S 100) of opening the mold 3, the cooking membrane 10, and more particularly the inflation chamber 11, can be placed at an internal pressure which is substantially equal to the ambient atmospheric pressure P atm, to allow the cooking membrane to regain, by elastic return, its rest dimensions, or can even be temporarily placed under depression at an internal pressure strictly lower than the ambient atmospheric pressure P atm, by connecting the exhaust circuit to a suction or vacuum device, so as to force the radial contraction of the cooking membrane 10 towards the central axis Z3, as illustrated in FIG. 2, and this in order to minimize the radial size, and more generally the volume, of said cooking membrane 10 to facilitate the extraction of the bandage 2.

[0063] By simple convention, it may be considered that one is in a so-called “simple” deflation state when the internal pressure is between -0.02 bar (minus 0.02 bar) and +0.02 bar (plus 0.02 bar) in the inflation chamber 11, and that one is in a deflation state assisted by a vacuum when said internal pressure is less than or equal to -0.3 bar (minus 0.3 bar), for example between -0.5 bar (minus 0.5 bar) and -0.3 bar (minus 0.3 bar).

[0064] Note that, unless otherwise stated, pressure values ​​are given as relative pressure relative to ambient pressure, which is considered zero. Thus, in particular, a relative pressure of -1 bar (minus one bar) represents absolute vacuum.

[0065] Following the step (S 100) of opening the mold 3, and more preferably after the cooked bandage 2 has been removed from said open mold 3, the method according to the invention comprises a test step (S200).

[0066] This test step (S200) makes it possible to test the tightness of the cooking membrane 10, in order to detect a possible leak, and in particular a slight leak linked to a small perforation.

[0067] Said test step (S200) comprises a partial re-inflation sub-step (S201), during which, while the mold 3 is in the open configuration, the opening of the inlet valve 21 is controlled in order to introduce inflation fluid into the inflation chamber 11 to increase the internal pressure prevailing in said inflation chamber 11 and thus cause a re-inflation of the cooking membrane 10, as can be seen in FIGS. 3 and 4, while measuring said internal pressure prevailing in said inflation chamber 11 by means of the pressure measuring device 25.

[0068] It will be noted that, advantageously, prior to this sub-step (S201) of partial re-inflation, the cooking mold 3 has been placed in the open configuration and the cooking membrane 10 has been positioned in the space in such a way that the re-inflation of the cooking membrane 10 can be carried out freely, in a volume of the cleared space, without the external surface 10 out of said cooking membrane encountering any obstacle which could mask a leak, and more particularly without the external surface 10 out of the cooking membrane 10 coming, under the effect of the movement of the cooking membrane 10 induced by the re-inflation, to bear against a portion of bandage 2 or against a portion of the internal wall 3_in of the mold 3.

[0069] Thus, during the partial re-inflation sub-step (S201), the cooking membrane 10 extends freely, under the effect of the pressure difference which exists between the internal surface 10_in, subjected to the inflation fluid, and the external surface 10 out of the cooking membrane, subjected to the pressure of the ambient atmosphere, until reaching a state of equilibrium which depends on this pressure difference and the intrinsic elasticity of said cooking membrane 10.

[0070] Advantageously, the fact that the cooking membrane 10 inflates freely makes it possible in particular to ensure that no perforation affecting the cooking membrane 10 will be obstructed, and, on the contrary, that any perforation will be dilated by the partial re-inflation operation, and therefore more easily detectable, so that the entire surface of the cooking membrane 10, between the two heels 12, 13, is effectively subjected to the sealing test.

[0071] Of course, if the inflation circuit 20 comprises an exhaust valve 22 which is separate from the intake valve 21, said exhaust valve 22 is preferably closed prior to the partial re-inflation sub-step (S201), so that the re-inflation operation is carried out while said exhaust valve 22 is in the closed state, and therefore prevents the inflation fluid from escaping from the inflation chamber 11.

[0072] The partial re-inflation sub-step (S201) is advantageously carried out under the control of the control unit 27.

[0073] The test step (S200) then comprises, immediately following the partial re-inflation sub-step (S201), a stabilization sub-step (S202), which is triggered when the internal pressure prevailing in the inflation chamber 11 reaches a predetermined threshold, called “test pressure” P test, which is chosen to be lower than the cooking pressure P eur, and according to which the inflation chamber 11 is isolated by controlling the closing of the intake valve 21.

[0074] It should be noted that the re-inflation is indeed “partial”, in that said re-inflation brings the internal pressure to a test pressure P test which is lower, and even much lower, than the nominal cooking pressure P eur.

[0075] Thus, the test pressure P test is chosen sufficiently low to prevent re-inflation from causing a dangerous tear or even bursting of the cooking membrane 10.

[0076] However, the test pressure value P test is of course chosen to be sufficiently high to enable any leaks to be highlighted, in particular by causing an expansion of the cooking membrane 10, and more particularly an elastic extension of the cooking membrane 10, which causes a widening of the possible perforation(s). In particular, the test pressure is of course strictly higher than the ambient pressure, here in practice the ambient atmospheric pressure, which prevails around the cooking membrane 10 and around the mold 3, and which is exerted on the external surface 10 out of the cooking membrane 10 when the mold is in the opening configuration.

[0077] For the above-mentioned reasons, the test pressure P test, which is used as a threshold for triggering the stabilization sub-step (S202) and thus the isolation of the inflation chamber 11, and thus for defining the end of the partial re-inflation sub-step (S201), is preferably equal to or greater than 30 mbar, preferably equal to or greater than 50 mbar, more preferably equal to or greater than 80 mbar. As indicated above, such a low limit value ensures sufficient re-inflation of the cooking membrane 10 to clearly indicate a possible leak.

[0078] On the other hand, said test pressure P test is less than or equal to 1 bar, preferably less than or equal to 350 mbar, more preferably less than or equal to 200 mbar, and even more preferably less than or equal to 130 mbar. As indicated above, such a high limit value in particular avoids a sudden tearing of the cooking membrane 10.

[0079] Particularly preferably, the test pressure P test may be between 30 mbar and 130 mbar, even more preferably between 80 mbar and 130 mbar, and for example equal to 120 mbar.

[0080] Advantageously, the leak detection according to the invention can therefore be carried out at relatively low pressure, in a particularly safe manner for operators and equipment, and with low energy expenditure.

[0081] It will be noted that, by usual convention, the values ​​of the internal pressure to which reference is made in the present application are expressed in relative values, that is to say they correspond to the pressure difference which exists between the interior of the inflation chamber 11, and therefore more generally the interior of the cooking membrane 10, on the one hand, and the exterior of the cooking membrane 10 on the other hand, that is to say the pressure difference which exists between the internal pressure which is exerted on the internal surface 10_in of the cooking membrane and the external pressure, of the ambient atmosphere, which is exerted on the external surface 10 out of the cooking membrane, when the cooking mold 3 is in the opening configuration.

[0082] In practice, the test pressure values ​​P test indicated above, and more generally the pressure values ​​indicated in this application, therefore correspond, unless otherwise stated, to relative pressure values ​​expressed in relation to the ambient atmospheric pressure of the location in which the installation 1 is located.

[0083] As soon as the internal pressure reaches the value of the desired test pressure P test, the isolation of the inflation chamber 11 is triggered, so as to separate said inflation chamber 11 from the inflation circuit 20, which amounts to confining the inflation fluid present in the inflation chamber 11 inside said inflation chamber 11, and to inhibiting the regulation of the internal pressure, and therefore to allowing the internal pressure to evolve freely within the captive volume of the cooking membrane 10, and more particularly of the inflation chamber 11.

[0084] It will be noted that, strictly speaking, the captive volume concerned by the isolation includes not only the volume of the inflation chamber 11 itself, but also the portion of the inflation circuit 20 which remains in communication with the inflation chamber 11, so that said captive volume extends from the downstream outlet of the intake valve 21 to the upstream inlet of the exhaust valve 22, passing through, and including, the volume of the inflation chamber 11. However, for simple convenience of description, reference may be made to the volume of the inflation chamber 11 to designate this captive volume.

[0085] According to the timing sub-step (S202), a time delay of a predetermined duration d_202, called the “time delay period” d_202, is applied and, at the end of said time delay period d_202, the value called the “reference pressure” P ref of the internal pressure prevailing in the inflation chamber 11 is measured.

[0086] The triggering of the control of the intake valve 21 and the triggering of the time delay are advantageously simultaneous, and caused by the fact that the internal pressure crosses the threshold represented by the test pressure P test. In practice, the control unit 27 therefore simultaneously sends a closing command to the intake valve 21 and a start command to a timer programmed for the time delay period d_202, from the instant t1 when said control unit 27 has captured and processed the information from the pressure measuring device 25 according to which the internal pressure has reached the test pressure P test.

[0087] As indicated above, this delay period d_202 will correspond to a stabilization period, during which the system, in principle closed, stabilizes, which constitutes the inflation fluid which is captive in the inflation chamber 11 and which is left to itself, in the absence of active regulation, following the closing of the inlet valve 21.

[0088] In particular, the time delay will make it possible to defer the measurement of the reference pressure P ref after the passage and the attenuation, or even the extinction, of transient disturbances which result from the operation of closing the inlet valve 21.

[0089] For example, as can be seen in the graph of Figure 6, the time delay will make it possible to defer the measurement of the reference pressure P ref after the passage of a possible overshoot 28, that is to say a temporary overpressure relative to the threshold represented by the test pressure P test, overshoot 28 which results from the inertia due to the response time in closing the intake valve 21 as well as the response time of the pressure sensor 26 and the corresponding acquisition chain of the pressure measuring device 25, because these response times allow the very rapid pressure rise phase initiated by the partial re-inflation sub-step (S201) to continue.

[0090] Thus, the value of the reference pressure P ref which will actually be measured and used as a reference to detect possible leaks will correspond to a stabilized value, not polluted by transient disturbances such as overshoot or parasitic oscillations.

[0091] It will be noted that, in practice, the reference pressure P ref, as it is observable at the end of the time delay, and more generally at the end of the stabilization sub-step (S202), may, depending on the case, and in particular depending on the volume of the membrane and the temperature, be either lower, higher, or substantially equal to the setpoint represented by the test pressure P test. The representation of figure 6, in which the reference pressure P ref appears lower than the test pressure P test, is therefore not restrictive.

[0092] Preferably, the time delay period d_202 is chosen to be equal to or greater than 2 seconds, preferably equal to or greater than 3 seconds, and more preferably equal to or greater than 4 seconds.

[0093] Thus, the lower limit value of the duration of the time delay period d_202 is long enough to ensure that any possible delays have been allowed to pass. transient disturbances linked to the closing actuation of the inlet valve 21, to the reaction time of said inlet valve 21 and of the pressure measuring device 25, and where appropriate to the visco-elastic behavior of the cooking membrane 10, such as for example an excess 28 of the internal pressure beyond the test pressure value P test or damped oscillations of the internal pressure.

[0094] On the other hand, and in particular in consideration of one or other of the aforementioned low limit values ​​of the delay period d202, the delay period d202 is preferably also less than or equal to 60 seconds, preferably less than or equal to 30 seconds, more preferably less than or equal to 20 seconds, or even less than or equal to 10 seconds.

[0095] Thus, the upper limit value of the delay period is short enough not to unnecessarily prolong the stabilization sub-step (S202), and consequently not to unnecessarily delay the observation sub-step (S203) and therefore the actual leak search operation itself.

[0096] In practice, in most cases, the timeout period d_202 will be between 4 seconds and 30 seconds.

[0097] For example, the timeout period d_202 can be equal to 5 seconds.

[0098] Following the stabilization sub-step (S202), the test step (S200) comprises an observation sub-step (S203), during which it is observed how the internal pressure prevailing in the inflation chamber 11 evolves over a predetermined duration called the “observation period” d_203, in order to identify a change in the internal pressure, relative to the reference pressure P ref, which is representative of a leak situation.

[0099] The observation sub-step (S203) advantageously immediately follows the stabilization sub-step (S202), and therefore begins at time t2 when the measurement of the reference pressure P ref has been acquired, and which marks the end of the time delay period d_202 and of said stabilization sub-step (S202).

[0100] Preferably, the observation period d_203 allocated is equal to or greater than 5 seconds, preferably equal to or greater than 10 seconds, more preferably equal to or greater than 20 seconds.

[0101] Thus, the lower limit value of the duration of the observation period d_203 is sufficiently long so that a variation in the internal pressure, in particular a drop in the internal pressure, which is sufficiently significant to be detectable, can be obtained, even in the case where the leak rate is low, due for example to the smallness of the perforation affecting the cooking membrane.

[0102] On the other hand, and in particular in consideration of one or other of the aforementioned low limit values ​​of observation period d_203, the duration of observation period d_203 is preferably less than or equal to 5 minutes, preferably less than or equal to 3 minutes, more preferably less than or equal to 60 seconds, or even less than or equal to 45 seconds.

[0103] Thus, the upper limit value of the duration of the observation period d_203 is short enough not to unnecessarily prolong the duration of the test step (S200).

[0104] In practice, in most cases, the observation period d_203 will be between 5 seconds and 30 seconds.

[0105] Furthermore, the observation period d_203 will preferably be longer than the delay period d_202. Indeed, the transient disturbing phenomena that one seeks to cover by the delay period d_202 are generally quite fleeting, whereas one may need, during the observation sub-step (S203), a relatively longer observation time to obtain a pressure variation, attributable to a small leak, which is of sufficient magnitude to be significant and detectable.

[0106] For example, the observation period d_203 can be equal to 30 seconds, in particular following a delay period d_202 which will have been equal to 5 seconds.

[0107] The observation of the evolution of the internal pressure during the observation period d_203 can be carried out for example: - either by continuously measuring, and where appropriate recording, the internal pressure over time, from the instant t2 which marks the start of said observation period d_203, and by gradually comparing this instantaneous internal pressure with the reference pressure P ref, for example to detect the crossing of a pressure threshold or a pressure variation rate threshold, - either by taking a one-off measurement of the internal pressure, for example at time t3 which marks the end of the observation period d_203, and by comparing this one-off measurement with the value of the reference pressure P ref which prevailed at the start of the observation period, at time t2.

[0108] Whether continuous or punctual, the measurement of the internal pressure value is advantageously carried out by means of the aforementioned pressure measuring device 25, and more particularly by means of the pressure sensor 26.

[0109] As indicated above, during the observation sub-step (S203), we observe how the internal pressure prevailing in the inflation chamber 11 evolves over the observation period d_203 in order to identify a change in the internal pressure, compared to the reference pressure P ref, which is representative of a leak situation.

[0110] This analysis operation can in practice be carried out jointly, simultaneously or almost simultaneously, with the acquisition of the internal pressure measurement which is carried out during the observation period d_203.

[0111] The acquisition and processing of internal pressure measurements are advantageously carried out by the control unit 27, which has a computer programmed for this purpose.

[0112] The analysis operation makes it possible in particular to identify, among several scenarios of evolution of the internal pressure which will have been predetermined, a scenario which is indicative of a leak, and thus to detect the presence of a leak.

[0113] More preferably, the analysis operation may make it possible to identify the origin of the leak, and in particular to determine whether said leak affects the cooking membrane 10 or the inlet valve 21, that is to say to characterize the leak, depending on the internal pressure evolution scenario that will have been observed.

[0114] The analysis operation, and therefore the identification of the scenario, may be based on different parameters, such as for example the speed of variation of the internal pressure per unit of time, or the crossing (upward or downward), by the internal pressure, during the observation period d_203, of a pre-established pressure threshold, etc.

[0115] Preferably, the magnitude and sign of the internal pressure variation observed during the observation period d_203 and beyond are calculated and monitored for this purpose. particularly at the end of said observation period d_203, compared to the reference pressure P ref.

[0116] According to a preferred implementation possibility, applicable moreover whatever the leak scenario or scenarios which are the subject of the observation sub-step (S203) and the analysis operation, during the observation sub-step (S203), the internal pressure called “residual pressure” P resid which prevails in the inflation chamber 11 at the end of the observation period d_203, here therefore at time t3, is measured and the difference Delta P, called “deviation” Delta P, between said residual pressure P resid and the reference pressure P ref is calculated in order to determine the sign and the amplitude of the observed change: Delta P = P resid - P ref

[0117] Thus, in a particularly simple way, we calculate the algebraic difference between, on the one hand, the residual pressure P resid, that is to say the internal pressure measured at time t3 when the observation period d_203 ends, and on the other hand the reference pressure P ref, that is to say the internal pressure measured previously, at time t2 when the time delay period d_202 ended and the observation period d_203 began.

[0118] This difference advantageously provides two pieces of information: the sign of the pressure change, negative for a drop, positive for an increase, and the amplitude of the deviation observed compared to the reference pressure P ref. This deviation, reduced to the elapsed time, here equal to t3 - t2, between the measurement of the residual pressure P rès and the measurement of the reference pressure P ref, advantageously makes it possible to characterize the absence or presence of a leak flow, and where appropriate the origin and / or magnitude of said leak flow.

[0119] It should be noted that, as a variant, the internal pressure could be continuously monitored during the observation period d_203, the difference between the instantaneous internal pressure and the reference pressure P ref calculated in real time, and the presence of a leak concluded as soon as this difference exceeds, in absolute value, a predefined alert threshold, even before the observation period d_203 has ended.

[0120] However, the total duration chosen for the observation period d_203 being relatively short, the time saving provided by such a variant would not be significant, especially considering the time required to replace the cooking membrane 10 when a leak is detected.

[0121] It is therefore preferable, to save on measurement and calculation operations, to opt for a simple one-off reading of the residual pressure P resid at time t3 which marks the end of the observation period d_203, rather than for continuous monitoring.

[0122] According to a preferred implementation possibility, during the observation sub-step (S203), a drop in the internal pressure is observed over the observation period d_203, said drop being of an amplitude, in absolute value relative to the reference pressure P ref, which is greater than a predetermined threshold called “drop threshold” P drop thresh, and it is concluded that there is a leak situation in the cooking membrane 10.

[0123] In other words, a first development scenario is identified here which is indicative of a leak situation where said leak affects the cooking membrane 10.

[0124] This first scenario, noted "scenario 1" and represented in dashed lines in Figure 6, is characterized here by a Delta P deviation, that is to say an algebraic difference Delta P = P resid - P ref mentioned above, whose sign is negative and whose amplitude is, in absolute value, greater than the drop threshold P drop thresh.

[0125] If this first scenario is detected, a membrane failure alert is triggered, here by means of the control unit 27, in order to cause the production of bandages to be stopped by the cooking installation 1, and the defective cooking membrane 10 to be replaced.

[0126] Preferably, the drop threshold P drop thresh is, in absolute value, between 10 mbar and 150 mbar, preferably equal to 10 mbar.

[0127] Alternatively, or equivalently, the drop threshold P drop thresh may be defined as a fraction of the reference pressure P ref, which may, for example, be useful if it is difficult to reproducibly control the reference pressure P ref from one test to another, because the said reference pressure P ref tends to fluctuate from one test to another even though the value of the test pressure P test chosen is identical for these different tests. Thus, for example, the drop threshold P drop thresh may be chosen in a range of values ​​between 5% and 20% of the reference pressure reference P ref, for example between 8% and 10% of the reference pressure P ref; in concrete terms, this may amount to setting a drop threshold P drop thresh of the order of 10 mbar to 12 mbar with respect to a measured reference pressure P ref of the order of 120 mbar.

[0128] According to another preferred implementation possibility, which may be complementary to the previous one, during the observation sub-step (S203), an increase in the internal pressure is observed, said increase being of an amplitude, in absolute value relative to the reference pressure P ref, which is greater than a predetermined threshold called “rise threshold” P rise thresh, and it is concluded, during the analysis sub-step (S204), that there is a leak situation in the intake valve 21.

[0129] In other words, a second development scenario is identified here which is indicative of a leak situation where said leak this time affects the intake valve 21. Indeed, such a second scenario is indicative of a failure of the intake valve 21 which, if not properly closed, allows the inflation fluid under pressure to continue to penetrate inside the cooking membrane 10, in the inflation chamber 11.

[0130] This second scenario, noted as "scenario 2" and represented by a dotted line in Figure 6, is characterized here by a Delta P deviation, that is to say an algebraic difference Delta P = P resid - P ref mentioned above, whose sign is positive and whose amplitude is, in absolute value, greater than the increase threshold P rise thresh.

[0131] If this second scenario is detected, a valve failure alert is triggered, here by means of the control unit 27, in order to stop the production of bandages and to cause an inspection of the installation 1 and a maintenance operation of the inlet valve 21.

[0132] Particularly preferably, the increase threshold P rise thresh may be chosen equal to 0 (zero) mbar, insofar as it can be considered that the slightest increase in internal pressure which would occur, after the stabilization provided by the delay period d202, compared to the reference pressure P ref, can only result from a failure of the inlet valve 21.

[0133] Thus, according to a preferential possibility of implementation, we will be able to observe, over the observation period d_203, an increase in the internal pressure compared to the reference pressure P ref, and we will then conclude that there is a leak situation in the inlet valve 21.

[0134] Overall, the method, in this case the test step (S200), therefore provides, and makes it possible to identify, at least one scenario of evolution of the internal pressure which is indicative of a leak situation, and even more preferably provides (at least) two scenarios of evolution of the internal pressure which are each indicative of a leak situation, among: - a first scenario which is indicative of a leak affecting the cooking membrane 10, - a second scenario which is indicative of a leak affecting the inlet valve 21.

[0135] As indicated above, the identification of one or the other scenario, and therefore the diagnosis of the presence or absence of a leak, is preferably carried out by calculating the difference Delta P between the residual pressure P resid measured at the end of the observation period d_203 and the reference pressure P ref measured at the end of the delay period d_202, and by examining on the one hand the sign, positive or negative, of this difference Delta P, and on the other hand the amplitude of this difference Delta P, amplitude which can be compared for this purpose to the thresholds of fall P drop thresh and rise P rise thresh mentioned above.

[0136] Advantageously, the same test step (S200) can thus make it possible to verify both the sealing of the cooking membrane 10 and the sealing, and more generally the correct operation, of the intake valve 21; which intake valve 21 constitutes, where appropriate, the only isolation valve making it possible selectively to put the inflation chamber 11 into communication with the inflation circuit 20 or, on the contrary, to isolate said inflation chamber 11 from the inflation circuit 20.

[0137] Of course, if, according to a scenario called "normal evolution scenario", or where appropriate "third scenario", noted "scenario 3" and represented in solid lines in figure 6, we observe, during the analysis operation: - that the Delta P gap remains contained between the fall thresholds P drop thresh on the one hand and the rise threshold P rise thresh on the other hand, i.e. that said Delta P gap remains contained within the range [- P drop thresh; + P rise thresh], - or, equivalently, that the measured residual pressure P resid is between P ref - P drop thresh (lower limit) and P ref + P rise thresh (upper limit), then we conclude that there is no leak, and that it is possible to start a new cooking cycle with the same cooking membrane 10, without it being necessary, in particular, to replace said cooking membrane 10.

[0138] These different scenarios are illustrated in Figure 6.

[0139] After the observation sub-step (S203), the cooking membrane 10 is deflated, so that the internal pressure prevailing in the inflation chamber 11 has dropped to a level equal to, or even lower than, the ambient atmospheric pressure P atm.

[0140] To do this, the control unit 27 preferably places the inflation chamber 11 in communication, via the exhaust valve 22, first with an exhaust pipe which opens into the open air, at ambient atmospheric pressure P atm, then, to accelerate the end of the emptying process and assist in the retraction of the cooking membrane 10, with a vacuum circuit creating a depression, as described above.

[0141] Thus, the cooking membrane 10 returns to a retracted state, in which said cooking membrane 10 is pleated on itself, and which will allow the placement of a new raw bandage 2 in the mold 3 then the closing of the mold 3 for a new cooking operation.

[0142] Preferably, the inflation fluid used for the test step (S200) is a dry gas, that is to say a gas which has, when it reaches the inlet valve 21, and therefore more generally when it enters and occupies the inflation chamber 11, a specific humidity which is less than or equal to 0.005 kg / kg (five per thousand).

[0143] "Specific humidity" or "water content" means the ratio between the mass of water, in kilograms, contained in the gas in question and the total mass, in kilograms, of said gas (considered in the "wet" state, i.e. including said mass of water).

[0144] Advantageously, the use of a dry gas makes it possible in particular to avoid condensation phenomena likely to disturb the evolution of the internal pressure which prevails in the cooking membrane 10, and therefore makes it possible to avoid errors of interpretation. during the test step (S200), more particularly during the observation sub-step (S203) and the analysis operation.

[0145] Furthermore, using a dry gas as an inflation fluid eliminates the need to equip installation 1 with a water vapor generation and management circuit. This saves space and energy.

[0146] Preferably, it is the same dry gas, and therefore the same inflation circuit 20, which is used on the one hand for the test step (S200), and more particularly for the re-inflation sub-step (S201), and on the other hand for the cooking operation.

[0147] Preferably, the inflation fluid used for the test step (S200), and more preferably also for the cooking operations, is a gas containing more than 99.8% pure nitrogen.

[0148] Such a gas is advantageously commonly available in industry, and can be stored in the form of liquid nitrogen. Furthermore, this ensures that the gas is completely dry, with a water volume fraction of less than 40 ppmv (typically at 5 bar), and is neutral and non-hazardous.

[0149] Furthermore, it will be noted that, when the cooking installation 1 is provided with a transfer system which makes it possible, from the same side of the mold 3, to alternately unload the cooked bandage 2 from the mold 3 and then load a raw bandage 2 into said mold 3, replacing the cooked bandage, then the test step (S200) can advantageously be accomplished in masked time, in the time which separates the extraction of the cooked bandage 2 from the mold 3, which is in the opening configuration, and the reintroduction into said mold 3 of a new raw bandage 2. Indeed, the total duration of the test step (S200), including the time necessary for re-inflation, the time delay for stabilization, then the observation and analysis, is relatively short, typically between 40 seconds and 80 seconds.

[0150] In order not to excessively penalize the useful time of the cooking installation 1, the leak test step (S200) according to the invention will preferably not be carried out systematically at each cooking operation, but rather in a spaced-out manner, according to a predetermined periodicity.

[0151] Thus, preferably, the test step (S200) is applied according to a predetermined frequency, called the “test frequency”, which provides for spacing two consecutive test steps (S200), carried out on the same cooking membrane 10, of a plurality of cooking operations, carried out with said same cooking membrane 10, so that the impact of the duration of the test step (S200) on the useful time of the installation 1 is reduced by distributing on average this duration of the test step (S200) over the number of cooking operations of said plurality of cooking operations.

[0152] For example, the leak tests carried out in accordance with the test step (S200) will be spaced out by an interval of at least five cooking operations, at least ten cooking operations, or even fifty cooking operations.

[0153] In other words, the test frequency will preferably be chosen so that a test step (S200) is carried out every five, or at least five, cooking steps, preferably every ten, or at least ten, cooking operations, or even every fifty, or at least fifty, cooking operations.

[0154] More specifically, we could therefore have, for example, a test frequency equal to one test every fifty cooking operations, or one test every ten cooking operations, or equal to one test every five cooking operations.

[0155] Preferably, the test frequency increases with the number of cooking operations carried out by the same cooking membrane 10 considered, according to a predetermined planning law, so that said cooking membrane 10 is subjected to a test step (S200) all the more frequently as the number of cooking operations that said cooking membrane 10 has carried out increases.

[0156] In other words, the older the cooking membrane 10 becomes, the more frequently its leak-tightness is tested. Thus, for example, the testing frequency can be set at one test every fifty cooking operations for a new cooking membrane, then reduced to one test every ten cooking operations, then to one test every five cooking operations.

[0157] Furthermore, it is preferably carried out, as is visible in Figure 7, in masked time during a cooking operation, and more preferably during each operation cooking, while the mold 3 is in the closed configuration and while the cooking membrane 10 is subjected to the cooking pressure P eur, for example in the two to four minutes preceding the end of the cooking operation, a pre-diagnostic operation according to which the inflation chamber 11 is isolated by controlling the closing of the inlet valve 21 and the internal pressure prevailing in the cooking membrane is monitored in order to be able to detect an abnormal drop in the internal pressure indicative of a potential leak, and, in the event of detection of such an abnormal drop in the internal pressure indicative of a potential leak during the cooking operation in progress, the start of the next cooking operation is made conditional on the performance of a test step (S200).

[0158] In particular, it is thus possible, if necessary, to force the triggering of a test step (S200) immediately after the mold opening step (S100) which marks the end of the cooking operation during which the pre-diagnostic operation detected a potential leak, regardless of the initially planned test frequency.

[0159] Thus, if a warning sign of a leak is detected during the cooking operation, and therefore advantageously in hidden time, then a thorough leak test is triggered in advance, in accordance with the test step (S200), as soon as the mold 3 is reopened, in order to check with certainty the leaktightness of the cooking membrane 10 before any resumption of a new cooking operation. Of course, if the test step (S200) confirms the presence of a leak, then the installation 1 will be stopped until it is repaired, and in particular, if necessary, until the used cooking membrane 10 is replaced. Priority is thus given to safety and quality of production.

[0160] Such an arrangement makes it possible in particular to guarantee, without degrading the useful time of installation 1, an acceptable level of risk, despite the spacing of the test stages (S200) in accordance with the test frequency, and thus to obtain an optimized compromise between production quality and production rate.

[0161] The pre-diagnostic operation can advantageously be triggered while a cooking stage 30 is being completed during which said internal pressure has been maintained at the level of a desired cooking pressure P eur, as can be seen in FIG. 7.

[0162] As an indicative example, over a cooking operation duration of eight minutes, it will be possible, in accordance with the pre-diagnostic operation, to isolate the inflation chamber 11, that is to say, inhibit the pressure regulation, three minutes before the end of the cooking operation, while an internal pressure of the order of 16 bar to 21 bar prevails, and observe the evolution of the internal pressure. If a drop in the internal pressure of more than 80 mbar is observed, then a suspected leak will be concluded, and the triggering of the test step (S200) will be forced before starting a new cooking operation, for example directly at the end of the cooking operation in progress, when the mold 3 is reopened.

[0163] It will be noted that carrying out the pre-diagnostic operation towards the end of the baking operation advantageously makes it possible to ensure that the temperature of the mold 3 is stabilized, and in particular that the temperatures of the mold 3, the bandage 2 and the baking membrane 10 are stable, and that consequently any variation in internal pressure, and more particularly any drop in internal pressure, will indeed be attributable to a leak, and not to fluctuations in the temperature of the inflation fluid which is contained in the baking membrane 10, for example to fluctuations which would be due to a transfer of heat from the inflation fluid to the bandage 2.

[0164] Of course, the invention also relates as such to a cooking installation 1 for cooking bandages 2, said installation 1 comprising: - a mold 3 which is arranged to alternately adopt a closing configuration, in which the internal wall 3_in of said mold delimits a molding cavity corresponding to the negative imprint of the bandage 2 and allows a baking operation of said bandage, and an opening configuration making it possible to extract, after the baking operation, a baked bandage 2 from said mold and to replace it with a raw bandage 2, - a cooking membrane 10 whose internal surface 10_in delimits an inflation chamber 11 allowing the cooking membrane 10 to be inflated, and whose external surface 10 out is arranged to, during the cooking operation, and under the action of a pressure called "cooking pressure" P eur which is applied in the inflation chamber 11, come to bear against the internal surface 2_in of the bandage in order to press the bandage 2 against the internal wall 3_in of the mold 3, - an inflation circuit 20 which comprises at least one inlet valve 21 making it possible to introduce into the cooking membrane 10, in the inflation chamber 11, an inflation fluid under pressure in order to inflate the cooking membrane 10, - a pressure measuring device 25 arranged to measure the pressure called “pressure internal”, which prevails inside the cooking membrane 10, in the inflation chamber H, said installation 1 comprising a control unit 27 which is arranged to make said installation execute: - a step (S 100) of opening the mold 3, during which, after having deflated the cooking membrane 10, the mold 3 is placed in the opening configuration, - a step (S200) of testing in accordance with the leak detection method according to any one of the possibilities described above.

[0165] Of course, the invention is in no way limited to the embodiment variants described above, the person skilled in the art being able in particular to isolate or freely combine one or other of the aforementioned characteristics, or to substitute equivalents for them.

Claims

CLAIMS 1. Leak detection method intended to detect a leak within a cooking installation (1) for cooking bandages (2), said installation comprising: - a mold (3) which is arranged to alternately adopt a closing configuration, in which the internal wall (3_in) of said mold (3), which corresponds to the negative imprint of the bandage (2), delimits a molding cavity and allows a baking operation of said bandage, and an opening configuration which allows, after the baking operation, a baked bandage to be extracted from said mold (3) and replaced by a raw bandage, - a cooking membrane (10) whose internal surface (10_in) delimits an inflation chamber (11) allowing the cooking membrane (10) to be inflated, and whose external surface (10 out) is arranged to, during the cooking operation, and under the action of a pressure called "cooking pressure" (P eur) which is applied in the inflation chamber (11), come to bear against the internal surface (2_in) of the bandage in order to press the bandage (2) against the internal wall (3_in) of the mold, - an inflation circuit (20) which comprises at least one inlet valve (21) allowing the introduction into the cooking membrane (10), in the inflation chamber (11), of an inflation fluid under pressure in order to inflate the cooking membrane (10), - a pressure measuring device (25) arranged to measure the pressure, called “internal pressure”, which prevails inside the cooking membrane (10), in the inflation chamber (11), said method comprising: - a step (S 100) of opening the mold, during which, after having deflated the cooking membrane (10), the mold (3) is placed in the opening configuration, - a test step (S200) which comprises: a partial re-inflation sub-step (S201), during which, while the mold (3) is in the open configuration, the opening of the inlet valve (21) is controlled in order to introduce inflation fluid into the inflation chamber (11) to increase the internal pressure prevailing in said inflation chamber (11) and thus cause re-inflation of the cooking membrane (10), while measuring said internal pressure prevailing in said inflation chamber by means of the pressure measuring device (25), then a stabilization sub-step (S202) which is triggered when the internal pressure prevailing in the inflation chamber (11) reaches a predetermined threshold, called "test pressure" (P test), which is chosen to be lower than the cooking pressure (P eur), and according to which on the one hand the inflation chamber (11) is isolated by controlling the closing of the inlet valve (21), and on the other hand a time delay of a predetermined duration (d_202), called "time delay period" (d_202), is applied, and, at the end of said time delay period (d_202), the value called "reference pressure" (P_ref) of the internal pressure prevailing in the inflation chamber (11) is measured, then an observation sub-step (S203), during which it is observed how the internal pressure prevailing in the inflation chamber (11) evolves over a predetermined duration called "time delay period" (d_202), ...), is measured, then an observation sub-step (S203), during which it is observed how the internal pressure prevailing in the inflation chamber (11) evolves over a prede observation” (d_203), in order to identify an evolution of the internal pressure,compared to the reference pressure (P ref), which is representative of a leak situation., 2. Method according to claim 1 characterized in that the time delay period (d_202) is chosen on the one hand equal to or greater than 2 seconds, preferably equal to or greater than 3 seconds, more preferably equal to or greater than 4 seconds, low limit value, and on the other hand less than or equal to 60 seconds, preferably less than or equal to 30 seconds, more preferably less than or equal to 20 seconds, or even less than or equal to 10 seconds, high limit value, for example equal to 5 seconds.

3. Method according to claim 1 or 2 characterized in that, over the observation period (d_203), a drop in the internal pressure is observed, said drop being of an amplitude, in absolute value relative to the reference pressure (P ref), which is greater than a predetermined threshold called "drop threshold" (P drop thresh), and it is concluded that there is a situation of leakage of the cooking membrane (10).

4. Method according to claim 3 characterized in that the drop threshold (P drop thresh) is comprised, in absolute value, between 10 mbar and 150 mbar, preferably equal to 10 mbar.

5. Method according to one of claims 1 to 4, characterized in that, over the observation period (d_203), an increase in the internal pressure is observed compared to the reference pressure (P ref), and it is concluded that there is a leak in the intake valve (21).

6. Method according to one of the preceding claims, characterized in that the observation period (d_203) allocated is on the one hand equal to or greater than 5 seconds, preferably equal to or greater than 10 seconds, more preferably equal to or greater than 20 seconds, low limit value, and on the other hand less than or equal to 5 minutes, preferably less than or equal to 3 minutes, more preferably less than or equal to 60 seconds, or even less than or equal to 45 seconds, high limit value, for example equal to 30 seconds.

7. Method according to one of the preceding claims, characterized in that, during the observation sub-step (S203), the internal pressure known as “residual pressure” (P resid) which prevails in the inflation chamber (11) at the end (t3) of the observation period (d_203) is measured, and the difference between said residual pressure (P resid) and the reference pressure (P_ref) is calculated in order to determine the sign and the amplitude of the observed change.

8. Method according to one of the preceding claims, characterized in that the inflation fluid used for the test step (S200) is a dry gas, that is to say a gas which has, when it reaches the inlet valve (21), a specific humidity which is less than or equal to 0.005 kg / kg, preferably a gas containing more than 99.8% pure nitrogen.

9. Method according to one of the preceding claims, characterized in that the test pressure (P test) which is used as a threshold for triggering the stabilization sub-step (S202) is on the one hand equal to or greater than 30 mbar, preferably equal to or greater than 50 mbar, more preferably equal to or greater than 80 mbar, lower limit value, and on the other hand less than or equal to 1 bar, preferably less than or equal to 350 mbar, more preferably less than or equal to 200 mbar, and even more preferably less than or equal to 130 mbar, upper limit value.

10. Method according to one of the preceding claims, characterized in that the test pressure (P test) is between 80 mbar and 130 mbar.

11. Method according to one of the preceding claims, characterized in that the test step (S200) is applied according to a predetermined frequency, called "test frequency", which provides for spacing two consecutive test steps (S200), carried out on the same cooking membrane (10), of a plurality of cooking operations, carried out with said same cooking membrane (10), so that the impact of the duration of the test step (S200) on the useful time of the installation (1) is reduced by distributing on average this duration of the test step (S200) over the number of cooking operations of said plurality of cooking operations, said test frequency preferably being chosen so that a test step (S200) is carried out every five, or at least five, cooking steps, preferably every ten, or at least ten, cooking operations, or even every fifty, or at least fifty, cooking operations.

12. Method according to claim 11 characterized in that the test frequency increases with the number of cooking operations carried out by the same cooking membrane (10) considered, according to a predetermined planning law, so that said cooking membrane (10) is subjected to a test step (S200) all the more frequently as the number of cooking operations that said cooking membrane has carried out increases.

13. Method according to one of the preceding claims, characterized in that a pre-diagnostic operation is carried out, in masked time during a cooking operation, and preferably during each cooking operation, while the mold (3) is in the closed configuration and while the cooking membrane (10) is subjected to the cooking pressure (P eur), for example in the two to four minutes preceding the end of the cooking operation, according to which the inflation chamber (11) is isolated by controlling the closing of the inlet valve (21) and the internal pressure prevailing in the cooking membrane in order to be able to detect an abnormal drop in internal pressure indicative of a potential leak, and, in the event of detection of such an abnormal drop in internal pressure indicative of a potential leak during the current cooking operation, the start of the following cooking operation is made conditional on the performance of a test step (S200).

14. Cooking installation (1) for cooking bandages (2), said installation comprising: - a mold (3) which is arranged to alternately adopt a closing configuration, in which the internal wall (3_in) of said mold delimits a molding cavity corresponding to the negative imprint of the bandage (2) and allows a baking operation of said bandage, and an opening configuration making it possible to extract, after the baking operation, a baked bandage (2) from said mold and to replace it with a raw bandage (2), - a cooking membrane (10) whose internal surface (10_in) delimits an inflation chamber (11) allowing the cooking membrane to be inflated, and whose external surface (10 out) is arranged to, during the cooking operation, and under the action of a pressure called "cooking pressure" (P eur) which is applied in the inflation chamber (11), come to bear against the internal surface (2_in) of the bandage in order to press the bandage (2) against the internal wall (3_in) of the mold (3), - an inflation circuit (20) which comprises at least one inlet valve (21) allowing the introduction into the cooking membrane (10), in the inflation chamber (11), of an inflation fluid under pressure in order to inflate the cooking membrane (10), - a pressure measuring device (25) arranged to measure the pressure known as "internal pressure", which prevails inside the cooking membrane (10), in the inflation chamber (11), said installation being characterized in that it comprises a control unit (27) arranged to make said installation execute: - a step (S 100) of opening the mold (3), during which, after having deflated the cooking membrane (10), the mold (3) is placed in the opening configuration, - a step (S200) of testing in accordance with the leak detection method according to any one of the preceding claims.

Images